Association of diet quality and body mass index in Mexican adults: a pseudo-panel analysis

Introduction: most studies that analyze the relationship between diet quality and obesity have a cross-sectional design; an alternative with repeated cross-sectional data is a pseudo-panel design. Objective: to estimate the association between trends in dietary patterns, defined by a diet quality index, and body mass index (BMI) of Mexican adults between 2006 and 2016. Methodology: a pseudo-panel analysis was performed using data from cross-sectional surveys: National Health and Nutrition Surveys of Mexico (ENSANUTs) 2006 and 2012 and the Midway National Health and Nutrition Survey 2016 (ENSANUTMC). Cohorts (n = 108) were constructed by grouping adults 20-59 years old by sex (men n = 6,081 and women n = 11,404), education level, and year of birth. The association between diet quality (defined with the Healthy Eating Index-2015) and BMI was estimated using a fixed effects model, adjusting for sociodemographic characteristics. Results: a one-point increase in the proportion of women with high diet quality was associated with 4.1 points lower BMI (p = 0.014) compared with women with low diet quality when excluding sub-reporters of energy, the same association is observed when physical activity is included in the model. No association was found between diet quality and BMI in men, possibly because of the existence of latent classes within sociodemographic strata, therefore diet qualiy is inversely associated with BMI only in some categories of sociodemographic strata. Conclusions: these results contribute to the evidence in the longitudinal analysis between diet and BMI, highlighting the importance of differentiating the population by sex and sociodemographic characteristics. These results are input for public policy creation that promotes improving the quality of the population's diet as part of multisectoral strategies to reduce overweight and obesity in Mexican adults.